Toward a deeper-rooted etymology of saplél

It should be a Sahaptian-family word, shouldn’t it, judging by where it was first documented?


(Image credit: Legends of America)

Alan Hartley’s good book, “Lewis & Clark Lexicon of Discovery”, indexes it under the Chinuk Wawa-oriented spelling < sapolil >, with variations from the L&C party’s journals including < cha-pel-el >, < shapelell >, < shap-e-leel >, < chappalell >, < Shapillele >, < Chapellell >, and < Shappallel >

Such variation in spelling isn’t necessarily of great significance, as both “sh” and “ch” could indicate a sound /š/ as in “shush”. In most languages of the region these sounds can alternate with a plain old “s”, which helps explain the form that the word eventually took in Jargon.

It’s my understanding that traditional < sapolil > are often made from biscuitroot, i.e. Lomatium species, i.e. the Sahaptian-etymology “cous(e)” … which may also have been an early CW word.

Some of the most compelling data that I’ve found for a Sahaptian-family origin of saplél are in the Nez Perce language.

There, I see a Causative prefix that varies in form, cap(á·) ~ cep(é·), due to requirements of vowel harmony with following word roots. This group of prefixes, like many Sahaptian affixes, perhaps originated from a verbal root, in this case ‘to press, to do something by pressure’. We’re told that the variant shapes without a final vowel refer to action on multiple objects. There also is a cognate Causative prefix sap(á·) ~ sep(é·), that lacks the sense of ‘pressing’.

Here are a couple of topically relevant example words with this prefix:

  • cap-k’í·cay ‘sun-dried paste of biscuit root’
  • capá·-k’imlaʔkt evidently = to form ‘smoked patties of biscuit root (qá·ws)’ by pressing

Let’s proceed from that information to investigate 3 or more separate etymological hypotheses…


And now to take you on a seeming tangent — in Nez Perce there is a suffix -u(·)y ~ -o(·)y ‘to have the property expressed by the preceding adjectives’. Examples include:

  • kíc-uy ‘metal; money; dollar; gold’, thus the same semantic range as CW chíkʰəmin, which may in fact be a (metathesized) source of the otherwise unknown Nez Perce root kíc, e.g. if Nez Perces did this thing they historically did — taking foreign words as being Salish, in this case entailing that CW chíkʰəmin involves the common Salish verbal transitivity suffix (of mental activities, perceptions, etc.) -min. Further evidence that NPerce kíc is a loan is that the root isn’t found in closely related Ichishkíin (Yakama) or Umatilla. 
  • k’áp-oy ‘mortar, basket mortar, hopper mortar; boxes made by hollowing a piece of log and used for grinding grains, roots, and so forth; hooked sticks to peg down ʔímse [‘mortar base’]’ — cf. k’áp ‘onomatopoeic word for (1) sound of hitting (e.g., a man with a club), pow, kapow…; (2) sound of biting teeth’
  • mil-ú·y ‘to be soft and damp (as bread soaked in liquid), be wet’

I don’t imagine that this suffix plays any role in the etymology of Chinuk Wawa’s saplél, but it’s important to be aware of it when we examine these next two example words:  

  • líl-uy ‘dirt, filth’
  • lil-ó·y ‘to be glad, be happy; to be thankful’, with an inflected form sapalló·yca ‘I make mine glad’ 

Both of those — also apparently unique to Nez Perce among the Sahaptian languages — have a root of the right shape to be involved in forming CW’s saplél. Of the two, I hazard a guess that the ‘glad’ one would be more likely to be used in a word for food! 

So maybe the origin of CW’s saplél is in a construction sep-/cep- ’cause (by pressing)’ líl ‘glad; thankful’. If that’s anything like an accurate view, I have to consider why biscuitroot loaves would ever be called such a thing. I can only point to the nature of those loaves as a crucial wintertime staple food. 


In Sahaptian languages, we have to always keep in mind the consonant sound symbolism that makes alternations like /l ~ n/ omnipresent. So our supposed lil should also be thought of as, potentially, nin. (We don’t seem to find words that include both sounds at once; there’s a kind of consonant harmony that occurs, so I haven’t found any shapes like *lin or *nil.) 

And there is a Nez Perce verbal suffix -in’ ‘passive participial’ that has the form -nin’ when it’s word-final. Could that suffix be part of CW saplél‘s etymology?

We’d have to postulate an unusually-formed NPerce word, containing only a prefix and a suffix! (Salish languages do that, but I don’t think Sahaptian ones do.) But as I noted above, there seems to be a historical pattern of Sahaptian prefixes originating from (previously) root words, so could it be that the sep-/cep- here were just that, which in turn would allow the use of the –nin’ suffix? 

The advantage of this hypothesis is its clear semantics, ‘multiple things caused to be pressed’, which is the action involved in forming these loaves. 


In Ichishkíin (Yakama) Sahaptin, there’s a noun root ninú ‘sand’. Would there be any reason to suppose that CW saplél traces back to a word for ‘pressing sand’?


We also have to take on the issue of why no such word as *seplíl / *ceplíl / *sepnín / *cepnín has been documented in Nez Perce! Was it an older term that became obsolete for some reason? 


This word is known in the Kiksht Upper Chinookan language, a neighbor of Sahaptian, and I understand biscuitroot loaves to have been an item of trade. Maybe this word is Chinookan, not Sahaptian, in origin. But my grasp of Chinookan morphology suggests to me (but doesn’t prove) that it’d be hard to form a word like saplél in Chinookan, which is why I’ve turned to Sahaptian instead. 

What do you think?
qʰáta mayka tə́mtəm?